{"title":"Comparison of Named Entity Recognition Tools Applied to News Articles","authors":"S. Vychegzhanin, E. Kotelnikov","doi":"10.1109/ISPRAS47671.2019.00017","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00017","url":null,"abstract":"Named Entity Recognition in texts is an important natural language processing task. There are many systems to solve this problem. These systems differ in targeting domains, processing methodologies, supported languages and recognized entity types. The presence of a large number of aspects creates difficulties for the user when choosing the appropriate tool for solving a specific problem. The aim of this work is a comparative study of seven publicly available and well-known libraries that can elicit named entities: Stanford NER, spaCy, NLTK, Polyglot, Flair, GATE and DeepPavlov. The article consists of seven sections. The introduction lists the areas of application for the Named Entity Recognition task and the approaches used to solve it. The second section is devoted to a review of works in which comparative studies of existing tools are presented. In the third section, the characteristics of the four text corpora that were used during the experiments are given. The fourth section contains a brief description of the tools selected for research. The fifth section describes the metrics used to evaluate tool performance. The sixth section presents the results of the experiments and their discussion. In conclusion the results of the work are summarized. The results of the study show that for the English language close values of the F1-score for the problem of Named Entities Recognition have the Flair and DeepPavlov libraries. For the Russian language the first place is taken by the DeepPavlov library, significantly surpassing other tools in quality.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125019857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The Study of Neurodynamic Systems of Continuous Adaptive Control","authors":"V. Sudakov, V. Osipov, Y. Nechaev, A. Vasilyev","doi":"10.1109/ISPRAS47671.2019.00027","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00027","url":null,"abstract":"A formal mathematical formulation of the problem of control and prediction of dynamic object behavior in interaction with the environment based on a neurodynamic system was proposed. Such point of view allowed us to consider from the system positions the conceptual schemes of building an integrated complex of multiprocessor computer learning environment of an interpreting neurodynamic system of continuous adaptive control of complex dynamic objects at the stages of their life cycle. The basis of methodological support are the principles of system analysis, dynamic catastrophe theory, methods and technologies of neuro-fuzzy and neuroevolutionary modeling, which allows formalizing the procedures of processing and interpretation of information about the environment and physical experiment data based on the achievements of modern intelligent technologies and high-performance computing.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129582362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Romanova, S. Strijhak, Konstantin Koshelev, M. Kraposhin
{"title":"Modeling the Reservoir Flooding Problem Using the Extended Version of the PorousMultiphaseFoam Library","authors":"D. Romanova, S. Strijhak, Konstantin Koshelev, M. Kraposhin","doi":"10.1109/ISPRAS47671.2019.00025","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00025","url":null,"abstract":"In this work, we develop software based on porousMultiphaseFoam library in OpenFOAM package to study enhanced oil recovery methods, and to study the influence of these methods on the oil recovery factor. New models of relative phase permeability and capillary pressure were implemented, and their comparative analysis was performed. The code was verified using the analytical approach. A new solver for simulating compressible flow was developed.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126505979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The Application of Modular Arithmetic for Matrix Calculations","authors":"V. Kuchukov, M. Babenko","doi":"10.1109/ISPRAS47671.2019.00013","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00013","url":null,"abstract":"The article considers the application of the Residue Number System (RNS) for matrix calculations. This can be useful to improve the performance and reliability of calculations in digital signal processing and artificial neural networks. We have reviewed the algorithms for calculating the scalar product. Their modeling and comparison with the standard means of the NumPy library of the Python are carried out. Methods of translation between the positional numeral system and RNS are described and simulated. Conclusions about the applicability of methods for calculations on Python are made. We have present the fault-tolerant scheme for error detection and correction. The examples of error occurrence and their correction by the presented scheme are considered.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124050569","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Gurin, A. Baykin, Timofey Polyansky, A. Krivtsov
{"title":"FEMEngine: Finite Element Method C++ Code Based on Functional and Template Metaprogramming","authors":"A. Gurin, A. Baykin, Timofey Polyansky, A. Krivtsov","doi":"10.1109/ISPRAS47671.2019.00020","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00020","url":null,"abstract":"The paper discusses the problems of a finite element method programming. Modern C++ functional programming and template metaprogramming approach to finite element analysis is presented. This approach simplifies the implementation of an effective assembly of the stiffness matrix for a problem defined by a weak form. This method is tested by a solution of the Poisson equation on an unstructured 3D tetrahedral mesh using FEM C++ library FEMEngine developed by authors. The function which calculates the matrix is generated by higher order functions during the compilation stage. The performance of the computation is analyzed by studying of a disassembled code and by comparison with the popular open source FEM software.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"684 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131885269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Computation Efficiency of S-CLSVOF Method for Additive Manufacturing Applications","authors":"M. Khomenko","doi":"10.1109/ISPRAS47671.2019.00022","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00022","url":null,"abstract":"The S-CLSVOF method is extended for direct numerical simulation in additive manufacturing applications. It is implemented in the framework of the OpenFOAM® and is appended by thermocapillary force and the energy equation. A number of 2D test cases are used for the verification which involves modeling of a stationary isothermal droplet, the Marangoni-driven flows in a rectangular cavity and the two-phase free surface Marangoni-driven flows. The tests are designed to verify the phenomena peculiar to additive manufacturing applications. It is shown that having a special treatment method could get the accurate results in a reasonable time.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116250500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Deep Learning for Customs Classification of Goods Based on Their Textual Descriptions Analysis","authors":"Anastasiia Ryzhova, I. Sochenkov","doi":"10.1109/ISPRAS47671.2019.00014","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00014","url":null,"abstract":"The article is devoted to the customs classification of goods based on their textual descriptions. In our work we propose different machine and deep learning models for solving this task using two variants of datasets: with or without duplicates. Finally we discuss and compare the obtained results","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"313 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123850554","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"High-Performance Flexible Memory Allocators in Complex Projects","authors":"I. Trub","doi":"10.1109/ISPRAS47671.2019.00010","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00010","url":null,"abstract":"The article proposes the methodology of data analysis and code design solutions to improve the flexibility of custom allocator assignment for simple data types and instances of data type classes. The methodology is based on special wrappers of memory functions, collecting data by the execution of source code, and analysing allocation/release traces with subsequent patterns extraction. Then the corresponding custom allocator is chosen for each pattern found and implemented in some module of source code as a container for user data, thus replacing malloc/free. The article describes the usage of the proposed approach for LLVM (Low Level Virtual Machine), which is a well-known tool for building compilers. It is shown that the choice of appropriate allocator provides the improvement of LLVM-based compiler's performance. An appropriate custom allocator is more effective than malloc, in particular, linear allocator is the best for StringMap and DomTreeNodeBase, stack allocator - for BitVector and linked list allocator - for Buckets. Aggregate performance improvement reaches around 10%. The methodology is implemented in shaders' compiler, but can be used in any project that uses LLVM or generally performs memory requests more or less intensively.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"117 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133696921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Parametric Study of the Accuracy of OpenFOAM Solvers for the Oblique Shock Wave Problem","authors":"A. Bondarev, A. Kuvshinnikov","doi":"10.1109/ISPRAS47671.2019.00023","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00023","url":null,"abstract":"The article is devoted to a parametric study of the accuracy of four solvers of the OpenFOAM package. As a test problem, we consider the classical two-dimensional problem of the fall of a supersonic inviscid compressible flow onto a flat plate at an angle of attack. As a result, an oblique shock wave is formed before the start of the plate. The problem is considered when varying the defining parameters of the problem, such as the angle of incidence of the flow on the plate and the Mach number. As a result of the calculations, a solution was obtained for the class of problems, where this class is determined by the ranges of variation of the determining parameters. For all solvers in the class under consideration, deviation surfaces from the known exact solution are constructed. The resulting estimates can be useful for users and developers of solvers.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129031803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Numerical Experiments with the Coupled Ocean-Earth-Atmosphere Circulation Model and the Analysis of Decadal Variability of Its Main Physical Characteristic","authors":"N. Tuchkova, K. Belyaev, Gury Mikhailov","doi":"10.1109/ISPRAS47671.2019.00028","DOIUrl":"https://doi.org/10.1109/ISPRAS47671.2019.00028","url":null,"abstract":"The statistical analysis of ensemble numerical experiments with a sample about 50 members of the Earth System Model outputs of the Max Planck Institute for Meteorology (MPI-ESM) from Hamburg, Germany is analyzed. The set of fifty initial conditions is set up and the model runs for 30 years were integrated from those conditions. The main model physical characteristics such as temperature, sear level and ice coverage are obtained and analyzed. Also, the most probable trajectories using the statistical ensemble Monte-Carlo scheme are investigated. The similarity of the model results with the natural tendencies is also shown. In particular, the specific properties of nonlinear equations are analyzed and discussed. Numerical calculations were carried out on the Lomonosov-2 supercomputer of Lomonosov Moscow State University and Mistral of DKRZ.","PeriodicalId":154688,"journal":{"name":"2019 Ivannikov Ispras Open Conference (ISPRAS)","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124848399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}